The science

Peer-reviewed evidence — this is not speculative.

The idea that disease has a smell is old. What is new — and what makes NeuroNose possible — is a decade of rigorous, replicated, peer-reviewed work that has turned that observation into measurable chemistry. Here is the evidence base, in full, with sources.

Canine detection · 2025
70–98%
sensitivity / specificity

In a double-blind trial, two trained dogs discriminated dry sebum skin swabs from people with Parkinson's versus controls. The dogs identified 70% and 80% of Parkinson's samples and correctly ignored 90% and 98% of controls — including patients with other health conditions. Both results were highly significant (Fisher's exact test, p < 0.0001).

Rooney N, Trivedi DK, Sinclair E, et al. "Trained dogs can detect the odor of Parkinson's disease." Journal of Parkinson's Disease 2025;15(6):1111–1115. DOI
Molecular basis · 2019
VOC
biomarkers identified

Mass spectrometry of sebum — hyphenated to a sniff-port so a hyperosmic "super-smeller" could correlate molecular peaks to the perceived odour — identified specific volatile compounds (including perillic aldehyde and eicosane) altered in Parkinson's. The anecdote became chemistry.

Trivedi DK, Sinclair E, Xu Y, et al. "Discovery of Volatile Biomarkers of Parkinson's Disease from Sebum." ACS Central Science 2019;5(4):599–606. DOI
Prodromal detection · 2025
Pre-PD
prodromal signal detected

The sebum volatile profile of people with isolated REM-sleep behaviour disorder — the single strongest prodromal marker of Parkinson's — sits intermediate between confirmed Parkinson's and healthy controls. The olfactory signal appears before clinical diagnosis.

Walton-Doyle C, Heim B, Sinclair E, et al. "Classification of Parkinson's disease and isolated REM sleep behaviour disorder…" npj Parkinson's Disease 2025;11:202. DOI
Why this is credible

Independent replication, not a single lucky result.

The strongest argument for an early-stage biomarker is convergence: different teams, different countries, different methods, all pointing at the same signal.

Canine detection of Parkinson's has now been demonstrated independently in the United Kingdom (Rooney 2025), China (Gao 2022 — three Belgian Malinois, sensitivity 89–91%, specificity 86–95%, published in Movement Disorders), and the United States (Holt 2024 — 23 breed-varied pet dogs, sensitivity 89%, specificity 87%).

In parallel, the molecular signature has been validated by metabolomics (Nature Communications 2021), a three-minute point-of-care mass-spectrometry assay (JACS Au 2022), electronic-nose systems (ACS Omega 2022), and live insect-brain biosensors (Biosensors & Bioelectronics 2024).

No single study is decisive. Together, they describe a real, measurable, reproducible olfactory biomarker — the foundation NeuroNose is built on.

The reference base

Full citations, grouped by evidence type.

Dogs detecting Parkinson's
2025
Rooney N, Trivedi DK, Sinclair E, Walton-Doyle C, Silverdale M, Barran P, Kunath T, Morant S, Somerville M, Smith J, Jones-Diette J, Corish J, Milne J, Guest C. "Trained dogs can detect the odor of Parkinson's disease." Journal of Parkinson's Disease 2025;15(6):1111–1115. 10.1177/1877718X251342485
Double-blind; 2 dogs; 60 control + 40 Parkinson's swabs; sensitivity 70–80%, specificity 90–98%; p < 0.0001. Consortium: Medical Detection Dogs, Bristol, Manchester, Edinburgh, Dundee.
2022
Gao C, Wang S, Wang M, et al. "Sensitivity of Sniffer Dogs for a Diagnosis of Parkinson's Disease." Movement Disorders 2022;37(9):1807–1816. 10.1002/mds.29180
Independent Chinese replication; 3 Belgian Malinois; medicated PD sensitivity 91%, specificity 95%; drug-naïve PD sensitivity 89%, specificity 86%.
2024
Holt L, Johnston SV. "From small to tall: breed-varied household pet dogs can be trained to detect Parkinson's Disease." Animal Cognition 2024;27:62. 10.1007/s10071-024-01902-5
23 breed-varied pet dogs; 4,553 trials; group-averaged sensitivity 89%, specificity 87%. Demonstrates trainability is not breed-restricted.
The sebum volatile-biomarker pipeline
2019
Trivedi DK, Sinclair E, Xu Y, Sarkar D, et al. "Discovery of Volatile Biomarkers of Parkinson's Disease from Sebum." ACS Central Science 2019;5(4):599–606. 10.1021/acscentsci.8b00879
Foundational study; identified perillic aldehyde and eicosane as principal volatile contributors to the Parkinson's odour signature.
2021
Sinclair E, Trivedi DK, Sarkar D, et al. "Metabolomics of sebum reveals lipid dysregulation in Parkinson's disease." Nature Communications 2021;12:1592. 10.1038/s41467-021-21669-4
Lipidomics extends the signature from volatile compounds to a broader lipid/metabolite class.
2022
Sarkar D, Sinclair E, Lim SH, et al. "Paper Spray Ionization Ion Mobility Mass Spectrometry of Sebum…" JACS Au 2022;2(9):2013–2022. 10.1021/jacsau.2c00300
Three-minute point-of-care mass-spectrometry assay from a skin swab — the precedent for a rapid reader.
2025
Walton-Doyle C, Heim B, Sinclair E, et al. "Classification of Parkinson's disease and isolated REM sleep behaviour disorder: delineating progression markers from the sebum volatilome." npj Parkinson's Disease 2025;11:202. 10.1038/s41531-025-01026-8
Prodromal evidence; iRBD sebum profiles intermediate between PD and healthy — detection before clinical diagnosis.
Electronic nose & machine olfaction
2022
Fu W, Xu L, Yu Q, et al. "Artificial Intelligent Olfactory System for the Diagnosis of Parkinson's Disease." ACS Omega 2022;7(5):4001–4010. 10.1021/acsomega.1c05060
Surface-acoustic-wave sensors + machine learning on sebum; the closest published precedent to the NeuroNose sensing device.
2024
Farnum A, Parnas M, Hoque Apu E, et al. "Harnessing insect olfactory neural circuits for detecting and discriminating human cancers." Biosensors and Bioelectronics 2024;257:116330. 10.1016/j.bios.2024.116330
Live insect-brain electrode arrays classify volatile signatures in ~250 ms — the biological precedent for neuromorphic olfaction.
The prodromal detection window
2008
Ross GW, Petrovitch H, Abbott RD, et al. "Association of olfactory dysfunction with risk for future Parkinson's disease." Annals of Neurology 2008;63(2):167–173. 10.1002/ana.21291
Honolulu-Asia Aging Study; impaired olfaction associated with 5.2× higher Parkinson's risk over 4 years.
2012
Doty RL. "Olfaction in Parkinson's disease and related disorders." Neurobiology of Disease 2012;46(3):527–552. 10.1016/j.nbd.2011.10.026
Canonical review; ≥90% of Parkinson's patients show measurable olfactory loss, often preceding motor symptoms.
2015
Devanand DP, Lee S, Manly J, et al. "Olfactory deficits predict cognitive decline and Alzheimer dementia in an urban community." Neurology 2015;84(2):182–189. 10.1212/WNL.0000000000001132
Olfactory deficits predict cognitive decline and conversion to Alzheimer's dementia.
2015
Berg D, Postuma RB, Adler CH, et al. "MDS research criteria for prodromal Parkinson's disease." Movement Disorders 2015;30(12):1600–1611. 10.1002/mds.26431
The international consensus criteria; olfactory loss is a formally recognized prodromal marker.
2019
Postuma RB, Iranzo A, Hu M, et al. "Risk and predictors of dementia and parkinsonism in idiopathic REM sleep behaviour disorder: a multicentre study." Brain 2019;142(3):744–759. 10.1093/brain/awz030
73.5% of iRBD patients converted to a defined neurodegenerative disease at 12-year follow-up.

What the evidence does — and does not — yet show

The Parkinson's evidence base is strong and replicated. The Alzheimer's evidence base is earlier-stage: olfactory dysfunction — a patient's own loss of smell, a separate phenomenon from the emitted signature NeuroNose detects — is an established Alzheimer's biomarker (Devanand 2015), but the specific volatile signature has not yet been characterized, and no peer-reviewed study has demonstrated animal detection of Alzheimer's from biospecimens. NeuroNose treats Parkinson's as the Phase 1 indication and Alzheimer's as a Phase 2/3 discovery program on the same platform. We would rather state this plainly than overclaim.

See how the science becomes a product.

Our technology →